Exploratory Data Analysis

Combining the datasets for 2017 and 2018 produces a sample of 128 community businesses, with up to 20 data points recorded for each1. However, the School for Social Entrepreneurs has not yet provided a complete data.

## Warning in mutate_impl(.data, dots, caller_env()): Unequal factor levels:
## coercing to character
## Warning in mutate_impl(.data, dots, caller_env()): binding character and
## factor vector, coercing into character vector

## Warning in mutate_impl(.data, dots, caller_env()): binding character and
## factor vector, coercing into character vector

Trading ratios

Performance

The average trading ratio in the Match Trading Group appears to rise by at least seven percentage points compared with a modest fall in Control Group. This is consistent with the core hypothesis of Trade Up, however there is considerable overalap between the two groups and the difference is not significant at the 5% level using traditional NHST-style tests.

Percentage point change
Group n Mean Median
Control 19 -1.8 -2.8
Match Trading 59 8.3 2.3 
## 
##  Kruskal-Wallis rank sum test
## 
## data:  Trading Ratio Change by Group
## Kruskal-Wallis chi-squared = 2.4513, df = 1, p-value = 0.1174
## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Trading Ratio Change by Group
## W = 426, p-value = 0.1188
## alternative hypothesis: true location shift is not equal to 0
Variable HDI Lower Bound HDI Upper Bound Mean Median Mode Credible % greater than zero
mu[1] -11.4 6.4 -2.6 -2.7 -2.8 NA
mu[2] -0.5 12.3 5.7 5.6 5.5 NA
muDiff -2.5 19.0 8.3 8.3 8.5 93.7
sigma[1] 9.8 25.5 17.5 17.1 16.6 NA
sigma[2] 13.5 27.6 20.6 20.5 20.9 NA
sigmaDiff -6.3 11.7 3.1 3.3 3.6 76.5
nu 1.2 49.6 13.8 7.0 3.8 NA
log10(nu) 0.3 1.7 0.9 0.8 0.7 NA
effSz -0.1 1.0 0.4 0.4 0.5 93.7

  1. Trading income and non-trading income, for each quarter, in the baseline year and programme year, plus four quarterly Trade Up grant awards.↩